LongGF: computational algorithm and software tool for fast and accurate detection of gene fusions by long-read...

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Long-read RNA sequencing (RNA-Seq) offers advantages over short-read RNA-Seq by capturing full-length transcripts.
• Existing gene fusion detection tools struggle with the high error rates and alignment complexities inherent in long-read RNA-Seq data.

Purpose of the Study:

• To develop a fast and accurate computational tool for detecting gene fusions from long-read RNA-Seq data.
• To address the limitations of current software in handling long-read sequencing data characteristics.

Main Methods:

• Development of LongGF, a C++ based computational tool for gene fusion detection.
• Evaluation using simulated long-read RNA-Seq datasets.
• Testing on real-world Nanopore direct mRNA, PacBio, and Nanopore cDNA sequencing datasets from cancer cell lines and acute myeloid leukemia samples.

Main Results:

• LongGF demonstrated superior performance in gene fusion detection on simulated datasets.
• The tool outperformed existing methods on Nanopore and PacBio sequencing data from cancer cell lines.
• LongGF precisely located a known translocation in acute myeloid leukemia at base resolution, validated by Sanger sequencing.

Conclusions:

• LongGF significantly enhances the discovery of gene fusion events from long-read RNA-Seq data, especially in oncology research.
• The tool's availability facilitates further research into gene fusions.
• LongGF provides a robust solution for analyzing complex genomic rearrangements from long-read sequencing.